The Baymard Institute: A glorious, evidence-based trove of UX best practices
What’s so great about Baymard and why aren’t there more organizations on the same level?
I realized I wanted to write this piece when I mentioned the Baymard Institute to a User Researcher with 10+ years of experience and they had no idea what I was talking about. They aren’t alone! I’ve gotten plenty of raised eyebrows on the subject before.
This is a shame!
If you’re a User Researcher (or in any area even tangentially related to UX, websites, or digital products) Baymard can probably provide you some value — quite possibly already has — and is worth being aware of.
I promise I’m not affiliated with them. I’m just a big fan!
The Baymard Institute is an “independent web usability research institute” founded by Denmark-based Christian Holst and Jamie Appleseed in ~2008.
Christian Holst and Jamie Appleseed
The organization was founded on the central thesis that design decisions are often made subjectively or politically, and that a more evidence-based approach is possible.
…and I know so because we work with a lot of the large fortune 500 and you think that they have this gigantic team that just knows exactly what’s going on in every single part of their site.
But when you then start to work with them it’s pretty clear that sometimes the reason why they have this particular design for their filters is because Mike decided so.
Mike is somewhere in the design department. [He] decided so and nobody argued against him and that’s why they have this filtering design — not because they spend hours and hours necessarily researching every single bit of that particular design.
— Christian Holst
Christian and Jamie also looked at academic research in the Human Computer Interaction field and noted that what they saw tended not to be as immediately commercially applicable as it potentially could be.
As an attempt at addressing this reality, Baymard sells research reports (and related services) generated based on tens of thousands of hours of their own in-house usability testing performed on industry-leading websites.
The full reports require payment (a la carte or through a subscription), but Baymard has also published more than 250 freely available articles on their site — each one containing a number of evidence-based recommendations.
Simply put, these articles are some of the best free resources out there for learning about concrete usability best practices. They’re one of my go-to recommendations for those new to UX design.
Is Baymard actually underrated?
Baymard is by no means unknown. In fact, they’ve been utilized by very recognizable brands such as Nike, Etsy, Walmart, and hundreds of others. But the recognition they get still seems to be out of proportion to the value they provide.
They have what I would consider a relatively small following on social media at less than 4,000 followers on Twitter and 1,500 on LinkedIn, and they don’t seem to be quite the household name among UX design and research circles that I would expect.
I think there are a few reasons for this.
One thing is that Baymard positions themselves as an e-commerce-specific research group. Those who don’t work in e-commerce may be quick to overlook the relevance of their reports and findings, even if many of their findings are actually widely applicable to other areas.
The narrower niche helps them connect better with brands they can provide the most value to, but this comes at the expense of communicating their value to that wider UX community.
It also doesn’t help that Baymard does almost no active marketing. They rely entirely on putting out free content and waiting for customers to come to them.
They’re a relatively small company (~15 employees) and don’t seem to have a particularly aggressive intent to grow or branch out beyond their core expertise.
I can’t say this approach hasn’t worked for them, but I would guess that they could grow their name recognition significantly if they experimented with more active promotion.
Examples of Baymard recommendations
You should definitely look through the full list of 200+ Baymard articles at some point if you haven’t, but to give you an idea of the type of things you’ll find, I’ll highlight a few here.
It should be pretty obvious that many of these best practices would have wide ranging applicability beyond just traditional e-commerce.
If you’ve ever worked on anything with a search function, settings/accounts page, homepage, form fields, mobile input, or anything else that could also be found in an e-commerce site, you should be able to find Baymard recommendations that are relevant.
2. 9 UX Requirements for Designing a User-Friendly Homepage Carousel (If You Need One)
3. E-Commerce Checkouts Need to Mark Both Required and Optional Fields Explicitly (Only 24% Do So)
4. Form Field Usability: Avoid Multi-Column Layouts (13% Get It Wrong)
5. Common Usability Pitfalls of Custom Designed Drop-Downs (31% Have Issues)
6. 3 Strategies for Handling Accidental ‘Taps’ on Touch Devices
7. How to Design ‘Applied Filters’ (42% Get It Wrong)
8. 6 Guidelines for Truncation Design
Unlike some puff-pieces you may commonly find online, each of these articles offer multiple specific, actionable, and evidence-based best practices.
The amount of research Baymard conducts to generate these recommendations is significantly higher than can be accomplished in individual, one-off studies — especially if you’re only testing limited non-code prototypes instead of fully functional websites or applications.
Their recommendations are also shown to be generally true across multiple products— meaning it isn’t likely that your users or use-cases are so unique that these findings would not at least somewhat apply.
Each article also comes with a built-in comment section for some additional clarifications, questions, and links. That willingness to expose themselves to dissenting opinions and alternative takes on their own platform is definitely appreciated.
I think Baymard is pretty great, but nothing is perfect.
You still need your own testing
Having access to general research doesn’t replace the need for usability testing your own product, though it does mean you can start your design work with smarter assumptions and spend less time rediscovering common issues.
Context is Key
When trying to understand a research-based recommendation it’s important to understand the context of the research. Ideally I’d prefer that the underlying raw data, such as timestamped video clips of test sessions, was made available.
Without this it can sometimes be a bit difficult to understand the severity or frequency of the issues that their recommendations are based on. I’ve purchased one-off reports before and still been left with some questions along those lines.
I haven’t ever had a full subscription to their service (working on it), so I can’t speak to the amount of context/services offered at that level. I think you do at least get access to some amount of direct user quotes and it’s possible there’s other additional context provided.
Also, Baymard has existed for 12 years, so you should pay attention to the date of each article as some of the recommendations may have been based on patterns that have fallen out of fashion. They’re generally good at updating things, but it’s something to be aware of.
EDIT: Check out Christian’s comment in the comment section below where he elaborates on Baymard’s update strategy & the additional context provided along with Baymard premium.
Why aren’t there more Baymards in the world?
Baymard represents more to me than just a repository of research.
In some ways it’s a fundamentally different way of thinking about how UX research gets done.
We spend a lot of time on doing studies that uncover things that have probably already been uncovered by other organizations dozens or hundreds of times before.
The reality is that very little work is perfectly original or unique. Commonalities exist across different products and product contexts that can be studied and turned into general best practices.
Centralizing the parts of research that can be centralized seems to make a lot of sense.
So why don’t we see Baymard-like research companies in areas other than e-commerce? The only one that seems to be remotely comparable is the more widely known Nielsen Norman Group (I’ll cover them at some point too!), but that’s pretty much it.
I can easily imagine a landscape where every distinct product category or subject (video streaming services, mobile apps, news content, enterprise tools/dashboards, accessibility, finance/insurance, etc) has their own Baymard-like research organization focused on uncovering the common best practices within that niche and selling the results.
There’s probably even room for multiple companies within the same category. But those companies don’t actually seem to exist. Why?
When asked, Christian has said he doesn’t know why more people aren’t doing what Baymard is.
There are some hurdles which are obvious enough:
- It can take a long time to build up enough research to the point that it’s valuable to start selling it.
- While you can gate access to reports you can’t actually copyright factual information. An organization may have issues dealing with their reports being pirated or their results being freely shared without them benefiting.
…but these are both issues that seem surmountable with the right business model.
While I myself don’t really know why there aren’t more Baymard Institutes, my suspicion is that there could be and it just hasn’t happened yet.
And if the opportunity exists then the only thing I’m left wondering is when someone’s going to take it.